African Educational Research Journal

Vol. 8(4), pp. 664-673, October 2020

DOI: 10.30918/AERJ.84.20.133
ISSN: 2354-2160
Full Length Research Paper

Gender and performance disparity in mathematics: A

study of South Western Uganda
Amos Musimenta, Francis Akena Adyanga* and Denis Sekiwu
Kabale University, P.O. Box 317, Kabale Municipality, Uganda.
Accepted 9 September, 2020

ABSTRACT

Gender has long been considered a factor contributing to differences in performance for male and female
students in diverse educational disciplines and levels. Although male and female students are taught in the
same classrooms in most Ugandan schools, there have been noticeable differences in Mathematics
performance in national examinations across the country. Thus, the aim of this study was to compare male
and female students’ performance in Mathematics and to establish factors accounting for the differences.
Using the Mixed method design, a sample size of 222 participants was recruited. The major findings
revealed that variation in Mathematics performance cannot be attributable to gender. The study
deconstructs the common gender-biased assumption that girls are naturally a ‘weaker sex’ and hence likely
to embrace subjects that are considered ‘soft’ such as language, literacy, communication skills, social
sciences among others. Such assumptions commonly fronted inadvertently without considering possible
negative consequences, are based on societal construction of social differences with no substantive
evidence as demonstrated in this study.

Keywords: Gender, mathematics, performance disparity, male, female.

E-mail: *Corresponding author. E-mail: [email protected].

INTRODUCTION

Gender refers to the social construction of diverse An example of such interventions is the formation of the
physical, biological, mental and behavioural Forum for African Women Educationalists (FAWE), a not-
characteristics relating to differences between the male for-profit Organization. The Organization was launched in
and female sex. It is a social and cultural construction of February 1997, as part of a Pan-African group with
roles, access to and control over resources between men strategies to promote gender equality in education on
and women, or boys and girls in society (Fennell and behalf of MoES. Uganda achieved gender parity on
Arnot, 2008; Chilisa and Ntseane, 2010; Fogliati and enrolment into primary schools in 2014. While in
Bussey 2013; MoES, 2016; Akena, 2020). Accordingly, Secondary schools, enrolment has increased from 54%
gender characterizes the differing social roles, for boys and 46% for girls in 2008 to 53% for boys and
responsibilities, constraints, opportunities and needs of 47% for girls in 2014. For Tertiary institutions, total
females and males in any given social context (Filgona, enrolment for females increased from 38% in 2002 to
2016; Akena, 2020). In Uganda, gender has long been a 44% in 2014 (MoES, 2016). Though there is increasing
common concept at the centre of the Ministry of evidence to show that the proportion of female students
Education and Sports (MoES) concern regarding taking up science courses in post-secondary institutions
academic performance. This concern mainly gained is on the rise, this rate is still lower among girls (Kaahwa,
ground between the 1990s and early 2000, with the 2012; Mbabaali, 2018; MoES, 2016).
advent of multiple education interventions to bridge Following the recommendations of the Education Policy
gender gaps in academic achievement (Kaahwa, 2012). Review Commission of 1998, the Ugandan government
 Musimenta et al. 665

in a bid to promote Science, Technology, Engineering in different geographical contexts that focuses on the
and Mathematics (STEM), introduced the Science comparison of performance in Mathematics examinations
Education Policy in 2005. The policy made science between boys and girls in different educational contexts.
subjects such as Physics, Chemistry, Biology and Mathematics is seen as a discipline that helps learners to
Mathematics obligatory for all secondary school students describe in concreate terms, ideas and relationships
up to form four (Adyanga, 2014). To boost enrolment in drawn from the environment (Kitta, 2014). Mathematics
science courses in post-secondary institutions, enables learners to make the invisible to be visible
government reserved over 70% of sponsorship funds in (Devlin, 2000; Patel and Dexter, 2014), thereby solving
tertiary institutions for science related courses. This is a difficult computational problems. However, there are
commendable action by the government because performance disparities in Mathematics among various
planning educational facility is not having roofs over the groups of students. Performance disparity refers to the
heads of students but providing the best possible differences in quality and quantity of knowledge, skills,
environment in which the formal and informal teaching- techniques, attitude, behaviour and philosophy that
learning process can take place (Musaazi, 2006). With different students acquire (Hanushek and Wößmann,
the above interventions to increase students’ science 2006; Muola, 2014), and the differences in performance
enrolment in secondary schools, the government also denote differences in student ability, which is evaluated
introduced the Secondary School Science and by marks and grades obtained through diverse modes of
Mathematics Teachers (SESEMAT) project aimed at assessment such as tests, examinations, coursework,
boosting the skill of science teaching for staff (MoES, among others (Nsubuga, 2008).
2016). SESEMAT, a three-year project launched in 2006, Male students’ academic achievements have
was made possible through partnership between Uganda traditionally been considered superior to that of female
government and Japan International Cooperation Agency students, especially in Mathematics and Science,
(JICA). On its part, JICA injected $1.5m USD to because of their higher levels of innate spatial abilities
SESEMAT pilot project with the aim of improving the (Benbow and Stanley, 1980) What is sometimes
teaching competencies of science teachers in secondary focussed on by researchers is performance in
schools. Much as these interventions have heightened Mathematics and science disciplines. The assumption is
enrolment in science courses, gender differentials that teachers may believe that males have higher levels
continue to manifest in academic performance especially of innate spatial abilities, which in turn affects the way
in Mathematics in secondary schools across the country. they treat males and females in the class (Benbow and
This study therefore examined how gender influences Stanley, 1982; Nosek and Smyth, 2011). Other studies
performance in Mathematics in Secondary Schools in have pointed out that female students generally perform
South Western Uganda. better in language due to their verbal and reasoning
To guide the audience, the rest of the paper is abilities (Wilberg and Lynn, 1999; Sahragard et al., 2011;
organized into four major parts as follows: Part one Jordan-Young, 2011). This, according to Erikson et al.
reviews literature in different geographical location with a (2012), can be attributed to the fact that infant girls are
focus on gender and academic performance. Part two encouraged to use more types of communicative
places the study within a fusion of qualitative and gestures than infant boys and both infant and toddler girls
quantitative research methodological pluralism. It argues produce more words than same-aged boys. Much as the
that methodological pluralism of data collection is the above literature posits that girls develop verbal and
best approach in obtaining rich information because reasoning abilities faster than boys which possibly has
through narratives, qualitative approach fills in any gaps later influence on academic performance between the
that cannot be articulated quantitatively. Part three two groups, other studies found some overlap in cognitive
discusses the use of qualitative and quantitative data development of the two (Grace et al., 2003). This links to
concomitantly. The discussion is interpolated with Petersen and Hyde (2014) who argued that gender
previously reviewed literatures to pull out convergences differences exist in academic ability and occupational
and divergences in quantitative data, participants’ oral goals mainly manifested in the field of science and
narratives, and perspectives of the researchers. And the Mathematics. Their rendition is that these gender
last part (part four) offers a conclusion with appeal to differences exist because of multiple factors such as
parents and guardians who are the key stakeholders in cultural anticipations, variations in developments,
education to take centre stage in ensuring children are stereotypes and prejudice from one context to another
given the right resources for education. (Petersen and Hyde, 2014).
Further, in Grace et al.’s (2013) study that looked at
contextual factors and alternative explanations for
LITERATURE REVIEW relationships between post-partum development (PPD)
and child cognitive development, their findings feed into
There is a growing body of literature based on research the research by Wilberg and Lynn (1999). Both of these
 Afr Educ Res J 666

studies revealed that boys of PPD mothers scored more this study because of its flexibility in generating data for
poorly on the perceptual, motor and verbal subscales of studying complex educational problems. Besides, using
the McCarthy Scales of Children’s Abilities. The the two research tactics allows strength of both
McCarthy Scales of Children's Abilities (MSCA) is a approaches to be combined, leading to a better
psychological test given to young children. The scales understanding of research problems that one approach
present carefully constructed individual tests to gauge the alone cannot provide (Creswell, 2002; Creswell and
child’s ability (Kaufman and Kaufman, 1973). In another Garrett, 2008; Akena, 2012). With this in mind, three
study, Eriksson et al. (2012) citing Maccoby and Jacklin methods of data collection were used, thus: interview,
(1974) argued that girls matured faster in verbal abilities questionnaires, and document analysis. The researchers
than boys. However, the study falls short of pronouncing examined documents (UNEB examination reports)
the reason for this faster maturity in verbal abilities obtained from the selected schools and in a range of nine
among girls. Responding to this bottleneck, scholars such years (2009 to 2018). Documents supplemented data
Jeje and Olajoke (2016) placed the blame on students’ from questionnaires and interviews conducted in June,
attitude towards Mathematics. They reasoned that male July and August 2019, respectively. Study participants
students have more positive attitude towards were recruited using purposive and convenient sampling
Mathematics than their female counterpart (cf, Hannula, techniques (as explained, for instance, by Onwuegbuzie
2002). Much as this may be the case, attitude alone and Leech, 2007).
cannot account for the performance disparity in The researchers chose four schools using convenience
Mathematics within Uganda education system. This is sampling based on their easy accessibility during the
because there are occasions when female students rainy season in the region since most roads are not
outperformed their male counterparts in Mathematics. tarmacked or maintained. Another reason for using
For instance, evidence from the 2018 Uganda convenient sampling strategy was the schools’ proximity
Advanced Certificate of Examinations (UACE) results to the researchers’ workstation (Adyanga and Romm,
shows that a total of 99,672 students registered for the 2016; Onwuegbuzie and Leech, 2007). Schools were
final examination. Out of this, 53,359 were males and also purposively sampled because of their status of being
41,313 were females. Female candidates performed amongst traditional schools with the largest students’
better at all levels with failure rate being low at 0.8% than enrolment in the region (Creswell, 2013). To protect
1.8% for male counterpart (Ahimbisibwe, 2019). In the confidentiality of the schools, pseudo names such
same year however, boys outsmarted girls in the lower Rwanda Senior Secondary (S.S), Angola S.S, Zambia
level of secondary school examination, the Uganda S.S and Kenya S.S were used. This is to ensure that no
Certificate of Education (UCE). According to the results, data get traced back to any single schools or participants.
boys performed better than their girls with 10.2% of the
boys passing in division one, compared to 6.6% in
division one for girls. Further, more boys (18.1%) passed Teacher recruitment
in division two compared to 13.9% for the girls (Kazibwe,
2019). It is this egregious representation of the scale of A homogenous sampling method was then used to
performance disparity between boys and girls that purposively select twenty-eight (28) teachers based on
motivated this study to gauge gender influence on the criterion of their teaching Mathematics in the selected
performance in Mathematics. Performance disparity schools (Adyanga and Romm, 2016). A few non-
between boys and girls is critically important when it Mathematics teachers were also conscripted in the study
comes to gender equity issues in occupational choices because of their specialty teaching sciences mainly
(Petersen and Hyde, 2014). Working with participant Physics, Chemistry and Biology. The researchers
voices from questionnaire responses and interviews, approached the schools and asked the headteachers
examination reports from Uganda National Examinations (after briefings about the study) for names of all
Board (UNEB), and published literatures, the study builds Mathematics teachers. From the list provided, the
on earlier research on gender and academic researchers contacted the teachers individually, briefed
performance. them about the study, and invited them to participate.
Some teachers declined participation claiming they did
not have time. To fill the gaps, selected teachers of
METHODOLOGY Physics, Chemistry and Biology were recruited. Those
who accepted to participate were given consent forms,
The study used a fusion of qualitative and quantitative which they signed and returned to the researchers.
research approaches to constitute the overall approach, Because some of the identified teachers were not at their
or what is called the mixed method research approach duty station on the day the researchers visited, their
(Tashakkori and Teddie, 2010; Hesse-Biber and consent forms were left in the office of the headteachers
Johnson, 2015). A mixed methods approach was used in after a telephone conversation with the researchers in
 Musimenta et al. 667

which their consents to participate were first orally questionnaire, interviews and documentary review as
sought. articulated below.

Recruitment of students Questionnaire

The second set of participants in the study was students This method was used to collect data from 194 students
from the four selected schools. Because these are from the four schools. After the consent for participation
vulnerable members (Belmont Report, 1979) of the was given on students’ behalf by the
community since all were below the age of 18 (in headteachers/deputy headteachers, students from each
Uganda, one is considered a child when he/she is below school were gathered in their Assembly Halls to meet
eighteen years old), the researchers sought consent for with the researchers. The researchers verbally briefed
their participation from their headteachers – since they students of the study objectives and how their
are in residential schools. When the headteachers participation can make a difference. From each of the
okayed students’ participation, they were given consent four schools, the researchers randomly picked students
forms to sign on behalf of the students. Therefore, all to participate in filling the questionnaires. For gender
students who participated in the study had their consent sensitivity, 100 students (50 boys and 50 girls) were
forms signed by either their headteacher or deputy picked from the first two schools, while 94 (44 boys and
headteacher. These two are the highest office bearer in 50 girls) were picked from the last two schools making a
the Ugandan school settings. The objective of the total of 194 participants under this category. Because of
research was then explained to the students and they the enthusiasm engendered about the study among this
were told that participation is voluntary, and they could group of participants, most of them filled and returned the
leave the study at any time. In the even that they chose questionnaires to their headteachers’ office the next day.
to quit the study, any information they had already Additionally, 28 teachers (12 females and 16 males) were
provided would be discarded. selected from the four schools to fill the questionnaire. In
total, two hundred twenty-two respondents (194 students
and 28 teachers) participated in filling the questionnaire.
Informed consent to teachers
The questionnaire itself was constructed as follows: It
Questionnaires were distributed in four secondary was taken from the Uganda Child and Wellness Survey
schools across Kabale District, South Western Uganda. and modified to the context of South western Uganda.
In the letter to potential participants in the teaching From this, we adapted the questionnaire to use a 5 Likert
category, the researchers explained that the central scale. The questionnaire contains five basic parts:
purpose of the study is to establish (using the teachers’ demographic information, gender issues questions, age
experiences in teaching Mathematics) the factors for issues questions, education background issues questions
continued examination performance disparity in and parents’ participation in supporting children with
Mathematics between boys and girls. As part of their homework.
participation, they are to fill the questionnaires and return
them to the researchers within a reasonable timeframe
possibly, a week. They were further informed that some Interviews
of them may later be invited to participate in face to face
interviews. The researchers further reaffirmed that their This method was used to obtain data from teachers and
participation in the study was voluntary and they have the headteachers. Upon successful signing of the consent
right to withdraw from the study any time without penalty forms, interview dates were set for the teachers who had
(cf. Adyanga and Romm, 2016; Romm, 2018; Schaefer accepted to participate. All the interviews took place
and Wertheimer, 2010; Sekiwu, 2013). Finally, the within the school premise either in the offices, staff rooms
researchers expressed that any information given in the or outside in the open ground under trees. Although all
questionnaires or during interviews (for those who would the 28 teachers were invited to participate in the interview
be invited to participate) will be kept confidential and session, only 17 (11 females and 6 males) honored the
quasi names would be used in the place of their names invitations. The interview questions were structured in a
and that of their schools. This is to ensure that no way that allowed for free conversations between the
information can be traced back to participants or schools. interviewers and participants with ample opportunities for
probing (cf. Adyanga and Romm, 2016). The following
were the key questions that guided the interview process:
Data collection methods
- What is gender and gender role?
Three methods of data collection were used thus - Elaborate what is meant by gender difference.
 Afr Educ Res J 668

- What factors account for the differences in performance contingent tables, t-statistic and chi-square. The tape-
in Mathematics between boys and girls in your school? recorded qualitative data were transcribed and then
- How does gender influence academic performance in analyzed for possible emerging themes arising from the
Mathematics in your school? participant voices. In the analysis process, the
researchers used qualitative data to back up findings
While the interviews were based on a set of obtained from the quantitative method of data collection.
predetermined questions, in order to create an In this way, descriptive data filled in gaps that could not
environment of trust, researchers avoided reading be explained quantitatively. For example, the quantitative
questions directly from the scripts but engaged in a data suggested that males performed much better than
conversational topic around the research theme gender females at Rwanda S.S in both O and A level
and its influence on performance in Mathematics. The examinations, but it was unclear why this might be the
conversations took place between each participant and a case. When we looked at the qualitative data, we were
member of the research team. In the process, the second able to suggest that the extremely large number of males
researcher listened and recorded selected responses of compared to female students at the school accounted for
respective participants, as well as noting down in a this quantitative finding.
notebook, any gestures and suggestive body language,
as these would provide useful insights during
interpretation of participant voices. RESULTS AND DISCUSSION
Also, a voice recorder was being run by the third
researcher (participants were informed that the The quantitative component of the study adopted a
conversation would be recorded to allow researchers multivariate approach to understanding the influence of
refer to the recorded files where clarity was later gender in performance in Mathematics. In particular, the
required). All participants agreed to their voices being researchers used contingent tables, t-statistic and chi-
recorded. Of the seventeen participants, twelve square to test for the existence of any significant
requested the researchers to provide them with the differences in Mathematics performance based on
recorded voices from their specific interview sessions. gender. The outcome of this approach is presented in
Two weeks after the study, the researchers honoured Table 1.
these requests by giving Compact Disks (CDs) to all The performance in Mathematics at UCE level
teachers who participated. The CDs contained recorded indicates that majority of the students scored credit five
interview sessions since they (participants) needed this (16.7%), of which 15.1% were male while 18.5% were
information for their private use. This was a way of female. This suggests that more female than male scored
thanking them for their participation in the study. a credit five. The percentage of students who scored a
distinction one (13.6%) was higher than those who
scored a nine (8.1%). Among the male, only 5.7% scored
Documentary review a nine compared to 10.9% among the female. This
suggests that more female than male failed Mathematics
A documentary review is a systematic process in which a
at UCE level. Contrastingly, more female (14.1%) than
researcher analyses the available literature in form of
male (13.2%) scored a distinction. The percentage of
reports and files for the purposes of retrieving information
female who scored a distinction one (13.6%) was more
relating to the subject under study (Bowen, 2009; Privault
than those who failed (10.9%). This suggests that
et al., 2012). For this study, the following documents
performance in Mathematics was better among female
were requested and received from the schools by the
than male. However, much as females performed better,
researchers: work plan, staff records of attendance,
the number of males taking up science related subjects in
examination results for UCE and UACE, scheme of work
high schools still outmatches the number of females
and lesson plans. The review of these documents
(Kaahwa, 2012; Mbabaali, 2018; MoES, 2016).
enabled researchers to acquire crucial information such
To paint a better picture on performance differences
as the number of times a teacher attended school in a
according to gender, the grades were grouped into levels:
semester, performance of students in national
1 – 2, distinction; 3 – 6, credit; 7 – 8, pass; and 9, failure.
examinations for the period under study, how often
Table 2 shows the details.
Mathematics lessons are taught in a week for each
Performance in Mathematics was grouped according to
school, etcetera. Such information became critical in
high performers that includes distinction and credit, and
buttressing the interpretation of quantitative data.
low performers for those in pass and failure. At distinction
and credit level, there were more girls (21.0%) than boys
Data analysis (19.7%) respectively. And these being higher grades, the
performance among girls appeared better than that of
Quantitative data were coded and analyzed using boys. Majority of the students passed with credit (42.8%)
 Musimenta et al. 669

Table 1. Performance in Mathematics at UCE level.

Mathematics at UCE
Gender
1 (%) 2 (%) 3 (%) 4 (%) 5 (%) 6 (%) 7 (%) 8 (%) 9 (%)
Male 13.2 9.4 6.6 13.2 15.1 13.2 9.4 14.2 5.7
Female 14.1 8.7 2.2 13.0 18.5 14.1 7.6 10.9 10.9
Total 13.6 9.1 4.5 13.1 16.7 13.6 8.6 12.6 8.1

Table 2. Grouped performance in Mathematics.

Math grade Total (%)

Gender High performers Low performers
Distinction (%) Credit (%) Pass (%) Fail (%)
Male 19.7 41.8 20.5 18.0 100
Female 21.0 44.0 17.0 18.0 100
Total 20.3 42.8 18.9 18.0 100

followed by those with distinctions (20.3%). This was this to insufficient teaching and learning aids. She
followed by those with pass (18.9%) and failure rate at succinctly stated
18.0%. Almost the same percentage of students scored a Our school has inadequate education facilities such as
pass and failed. The rendition from this table is that the Mathematic, English and Integrated science textbooks.
percentage of students who passed in the high When I took over leadership two years ago, this problem
improvement scale (distinctions and credits) favours the was compounded by high indiscipline cases among
females. While the male students are mainly dominant in students. With this, you cannot expect good academic
the lower improvement scale of pass and fail performance. Also, most of our students come from poor
respectively. Inferring from this table therefore, female socio-economic family background and cannot afford to
have high numerical competence than their male buy their own learning resources and often perform
counterparts which is a sharp contrast with most existing poorly in national examinations.
literatures which tends to put male above female The problem of poverty espoused by the above
(Benbow and Stanley, 1982; Kazibwe, 2019; Petersen participant was repeated with high intensity among most
and Hyde, 2014). teachers who partook in the different interview sessions
In Table 3, performance in Mathematics shifts direction across all the four schools. This also supported by
regarding school and gender. In Rwanda S.S for existing literatures which reasoned that schools in poor
example, there are more males at distinction level neighbourhood generally perform poorly due to scarce
(53.3%) than females (46.7%) while at credit level, there education resources (Kaahwa, 2012; Mbabaali, 2018;
are still more males (60.9%) than girls (39.1). This implies MoES, 2016, Awori et al., 2020). Among others, poverty
that males rank highest on the high-performing scale than is argued to limit parents’ ability to pay fees in time, buy
their female counterparts. More still, at the low performing school uniforms, feed children on balanced diets and buy
scale, males perform highest with pass (60%) and failure books and pens which influences academic
(71.4%) than females with 28.6% pass and 40% failure performances across subjects. In the words of a female
rates respectively. Although male and female registered teacher from Rwanda S.S, the picture is pithily painted
equal performance at distinction (50%) in Angola S.S, with the statement:
there is a slight alteration in the tide where more females
register more credits (61.9%) than males (38.1%). This … you can’t expect children to pay attention and
alteration in the performance tide continues to reflect grasp the content of any lesson when they are
more males (62.5%) with passes than females (37.5%). on empty stomach. Because we don’t have
However, the tide is short-lived where female perform school feeding programs, some of our students,
poorer (76.9%) than males (23.1%) as indicated by the especially those from poor family background
number of failures. In order to obtain some insight on why are often inactive during afternoon lessons due
this may be the case, we turned to some of the qualitative to hunger. They simply pass time in classes.
data. For instance, when asked to speak about the
rationale for poor performance among girls in Relating the above participant’s perspective to existing
Mathematics, a female headteacher at Angola S.S linked literatures, the challenge of poverty and child labour is
 Afr Educ Res J 670

Table 3. Mathematics performance according to school and gender.

Name of school Distinction (%) Credit (%) Pass (%) Fail (%) Total (%)
Male 53.3 60.9 60.0 71.4 60.0
Rwanda Female 46.7 39.1 40.0 28.6 40.0
Total 100 100 100 100 100

Male 50.0 38.1 62.5 23.1 40.0

Angola Female 50.0 61.9 37.5 76.9 60.0
Total 100 100 100 100 100

Male 63.2 56.0 60.0 66.7 59.7

Zambia Female 36.8 44.0 40.0 33.3 40.3
Total 100 100 100 100 100

Male 57.7 57.1 70.6 58.3

Kenya Female 100.0 42.3 42.9 29.4 41.7
Total 100 100 100 100 100

long known to be major factors accounting for children microscopic to account for wide discrepancies.
absenteeism from school with later negative influence on From Table 4, syllabus coverage (α = 2.459; sig. < .05)
academic performance (Duryea and Arends-Kuenning, indicates that Mathematics syllabus coverage at Uganda
2003; Emerson and Souza, 2008; Nielsen, 1998; Zhang, Certificate of Education (UCE) did not differ significantly
2003). This problem is not unique to any geographical across the sampled schools. All the schools gathered
context as most of our study participants think. It is a similar efforts to ensure syllabus coverage in
global challenge that requires multipronged interventions Mathematics, but it is not enough to use syllabus
of all stakeholders to ameliorate. coverage as a variable to gauge performance disparity
For Zambia S.S, males continue to uphold the tradition across schools. However, availability of textbooks (α =
whereby they had more distinctions (63.2%) than the 160.73; sig. < .05) and number of Mathematics lessons
female students (36.8%). The same continues to suffice (α = 83.459; sig. < .05) significantly varied across the
where more male had credits (56%) than females (44%). sampled schools. This is because different schools had
In Kenya S.S, the females took the day with 100% different quantities and quality of textbooks that would
distinctions and no male (0.0%). Conversely, the male help learners in improving their grades in Mathematics.
had more credits (57.7%) than their female counterparts Similarly, significant differences in the number of
(42.3%) while to another extreme, male performed poorer mathematics lessons per week suggest that different
(57.1%) with passes than the female (42.9%). This same schools accorded varying importance to the teaching of
trend continues with more male having failures (70.6%) Mathematics. Ideally, Mathematics is meant to be taught
than female (29.4%). This data feeds into the perspective in at least six lessons a week. However, the statistics
of the male headteacher from Kenya S.S who espoused denote that some of the schools provided less than six
that: Mathematics lessons and others more than the minimum
requirement. This probably explains the huge disparities
It is a mistake to tag academic performance in in Mathematics performance across the sampled schools
any subject to gender. Instead, there are many as lamented by a female participant from Zambia S.S
factors that account for differences and although during an interview session
gender could be one of them, it is at the lower Talking of factors for poor performance in Math in our
end of the scale. I have been at the centre of schools, the answer, to me, lie within. We don’t have to
leadership in high schools for 17 years and seen look further. The number one factor is absenteeism from
how the balance in performance keeps shifting duty among some of my colleagues. As a result, most of
between the two genders. them teach Math once a week instead of three times as
time tabled. In such a case, they do not finish the
The above participant and eight others therefore called curriculum and students go to final exams when half
for more studies to understand some of the hidden baked.
factors such as economic, environmental, and political The challenge of teacher absenteeism is not a preserve
influencing academic performance of students. Although to Zambia S.S but a general concern across the studied
they are not completely dismissive of the influence of schools as articulated by many participants during
gender, they argue that such influence is very interviews. It is also a general problem nationally. Except
 Musimenta et al. 671

Table 4. T-statistics.

Syllabus coverage Students' textbooks Number of Math lessons in a week

Chi-Square (a, b) 2.459 160.730 83.459
Df 2 2 2
Asymp. Sig. 0.292 0.000 0.000

for Angola S.S, teacher attendance records from the association.

sampled schools revealed that most of the teachers failed From the statistics above, there was no relationship
to complete the curriculum in the academic year concerning gender and performance in Mathematics
2017/2018 due to multiple absenteeism from work. among students in the schools investigated (α = 0.776;
Nationally, a study conducted by a Dutch agency SNV sig. ˂ .05). The implication here is that performance in
indicated that teacher absenteeism in Uganda is ranked Mathematics cannot not be attributable to gender
the highest in the world at 35% with teachers guaranteed disparity because there were instances when girls
to miss at least two days of work each week (Talemwa performed better than boys, and other instances where
and Eupal, 2009). We, therefore, believe that when boys took the lion’s share in performance. There are
schools fail to follow the Ministry of Education and Sports certainly multiple factors that needs critical examinations.
guideline to teach above the set minimum lessons, there For instance, a good grasp of language among girls as
is a likelihood for that school to register many failures in articulated in the literature (Wilberg and Lynn, 1999)
national examinations such as UCE and UACE. could contribute in one way or the other, to a better
The study also established if there existed any interpretation and understanding of examination
association between gender and performance in questions in Mathematics and other subjects leading to
Mathematics. Using cross tabulation procedures, chi- better performance (Sekiwu et al., 2020). This view is
square statistics were obtained (Table 5). Low Pearson widely shared by interview participants, many of whom
significant values indicate existence of some association disassociated academic performance in Mathematics
while high significant values indicate absence of from gender influence.

Table 5. Chi-squares statistics.

Value df Asymp. Sig. (2-sided)

Pearson Chi-Square 4.822 8 0.776
Likelihood Ratio 4.982 8 0.760
Linear-by-Linear Association 0.197 1 0.657

CONCLUSION disparities across the sampled schools between boys and

girls, Mathematics performance cannot be attributable to
As the preceding discussions illustrated, the gender gender. There are cases where boys outperformed girls
factor is considered important in Mathematics as well as instances where girls outpaced boys. The
performance since a significant portion of literature implications of this study to educational policy influencers
across the globe theorizes that there are performance are that the gender factor alone is inadequate in
variations between boys and girls. In order to promote accounting for the continuous performance disparity
educational equity, these performance variations must be between boys and girls in Mathematics across Ugandan
examined and ameliorated if detected. We also schools. This might have implications for the way in
recommend here, an epistemological grounding where which feminist theory is applied in an African context, as
theoretical declarations regarding the way gender is suggested by Chilisa and Ntseane (2014), where they
constructed in society, can be negotiated as “biased” and argue that we cannot transpose Western-generated
hence, needs to be problematized further. feminist theory into African contexts. It is therefore
This study, therefore, sought to examine the influence important that we constantly monitor and question the
of gender on Mathematics performance in secondary influence that social construction of gender, even from
schools in South Western Uganda. The findings of the African lived experience can have in society. This
study indicated that although there are performance process is critical for identification of potential leverage
 Afr Educ Res J 672

points for collective empowering actions. school and college of sex differences in mathematical reasoning
ability: A longitudinal perspective. American Educational Research
Therefore, the assumption that feminists across the
Journal, 19(4): 598-622.
board criticize — that girls are naturally a ‘weaker sex’ Bowen, G. A. (2009). Document analysis as a qualitative research
and hence likely to embrace subjects that are considered method. Qualitative research Journal, 9(2): 27-40.
‘soft’ such as language, literacy, communication skills, Chilisa, B., and Ntseane, G. (2010). Resisting dominant discourses:
Implications of indigenous, African feminist theory and methods for
social sciences among others are biased and based on gender and education research. Gender and Education, 22(6): 617-
societal construction of social differences (Akena, 2020). 632.
Also, the way in which these assumptions play out in Creswell, J. W. (2002). Educational research: Planning, conducting,
different social contexts cannot be generalized across all and evaluating quantitative (pp. 146-166). Upper Saddle River, NJ:
societies. In our study, such biased construction of social Prentice Hall.
Creswell, J. W. (2013). Research Design: Qualitative, Quantitative and
differences is debunked with evidences from the four th
Mixed Approaches (4 Edition). University of Nebraska—Lincoln. Los
mixed (boys and girls) schools with relatively large Angeles: Sage Publishers.
students’ enrolment. However, given the limited number Creswell, J. W., and Garrett, A. L. (2008). The “movement” of mixed
methods research and the role of educators. South African Journal of
of participants examined in this study, the findings cannot
Education, 28(3): 321-333.
in any way be claimed to represent the entire state of Devlin, K. (2000). The language of Mathematics: making the invisible
academic performance in Mathematics in the country. visible. Macmillan.
Instead, it should inspire more studies on larger scale to Duryea, S., and Arends-Kuenning, M. (2003). School attendance,
child labor and local labor market fluctuations in urban Brazil. World
derive data for transferability across the country, which in Development, 31(7): 1165-1178.
turn can contribute to the literature world-wide. Emerson, P. M., and Souza, A. P. (2008). Birth order, child labor, and
Finally, in keeping up with the principles of social school attendance in Brazil. World Development, 36(9): 1647-1664.
justice education, education policy makers should always Eriksson, M., Marschik, P. B., Tulviste, T., Almgren, M., Pérez Pereira,
cater for the gender and equity dimensions in schooling M., Wehberg, S., and Gallego, C. (2012). Differences between girls
and boys in emerging language skills: Evidence from 10 language
as a way of bestowing equal rights/access to quality communities. British Journal of Developmental Psychology, 30(2):
education for boys and girls. This feeds into the 326-343.
requirements of Uganda’s Vision 2040 that aim to Fennell, S., and Arnot, M. (2008). Decentring hegemonic gender
theory: The implications for educational research. Compare, 38(5):
establish a middle-class status for all Ugandans and to
525-538.
build an industrial society. For this to materialize, there is Fogliati, V. J., and Bussey, K. (2013). Stereotype threat reduces
need to invest in science education as well as ensuring motivation to improve: Effects of stereotype threat and feedback on
gender parity across technical and vocational training women’s intentions to improve mathematical ability. Psychology of
Women Quarterly, 37(3): 310-324.
institutions nationally. This cannot be accomplished by
Grace, S. L., Evindar, A., and Stewart, D. E. (2003). The effect of
the State alone. The family (parents and guardians) postpartum depression on child cognitive development and behavior:
needs to take a central lead as the first social institution a review and critical analysis of the literature. Archives of Women’s
naturally mandated with the upbringing and general Mental Health, 6(4): 263-274.
welfare of children. Hannula, M. S. (2002). Attitude towards mathematics: Emotions,
expectations and values. Educational Studies in Mathematics, 49(1):
25-46.
Hanushek, E. A., and Wößmann, L. (2006). Does educational tracking
REFERENCES affect performance and inequality? Differences‐in‐differences
evidence across countries. The Economic Journal, 116(510): C63-
Adyanga, F. A. (2014). African Indigenous Science in Higher Education C76.
in Uganda (Doctoral dissertation, University of Toronto (Canada). Hesse-Biber, S. N., and Johnson, R. B. (2015). The Oxford handbook
ProQuest LLC. of multimethod and mixed methods research inquiry. Oxford
Adyanga, F. A., and Romm, N. R. A. (2016). Researching Indigenous University Press.
science knowledge integration in formal education: interpreting some Jeje, O. S., and Olagoke, M. (2006). Mathematics attitude of male and
perspectives from the field. International Journal of Educational female students in junior secondary school in Ikole local government
Development, 3(1): 1-14. area, Nigeria. International Journal of Research in Education, 3(2):
Ahimbisibwe, P. (2019). Uganda: Girls Beat Boys in A-Level Exams. 191-194.
Retrieved from: https://www.monitor.co.ug/News/National/Girls-beat- Jordan-Young, R. M. (2011). Brain storm: The flaws in the science of
boys-A-Level-exams/688334-5004238-votef8/index.html. sex differences. Harvard University Press.
Akena, F. A. (2012). Critical analysis of the production of Western Kaahwa, J. (2012). The experiences of Ugandan females in
knowledge and its implications for Indigenous knowledge and Mathematics. Science Journal of Psychology, Volume 2012, Article
decolonization. Journal of Black Studies, 43(6): 599-619. ID sjpsych-103, 12 Pages.
Akena, F. A. (2020). The “Miss Curvy Uganda” pageant: representation, Kaufman, A. S., and Kaufman, N. L. (1973). Black-white differences at
commodification and exploitation of women’s bodies. Social ages 212− 812 on the McCarthy scales of children's abilities. Journal
Semiotics, 1-16. of School Psychology, 11(3): 196-206.
Awori, S., Denis, S., Frederick, S., and Frances, N. (2020). Ecology of Kazibwe, K. (2019). Boys outshine girls as 2018 UCE exams are
schooling: Enabling school environment for student engagement in released. Retrieved from: https://nilepost.co.ug/2019/01/31/boys-
Uganda’s Universal Secondary Education. International Journal of outshine-girls-as-2018-uce-exams-are-released/.
Educational Policy Research and Review, 7(2): 38-46. Kitta, S. (2014). Enhancing Mathematics Teachers’ Pedagogical
Benbow, C. P., and Stanley, J. C. (1980). Sex differences in Content Knowledge and Skills in Tanzania. Print Partners - Ipskamp:
mathematical ability: Fact or artefact. Science, 210(12): 1262-1264. Enschede.
Benbow, C. P., and Stanley, J. C. (1982). Consequences in high Mbabaali D. (2018). Why are girls still few in science-related courses?
 Musimenta et al. 673

Daily Monitor, Retrieved from Talemwa, M., and Eupal, F. (2009). Uganda teacher absenteeism is
https://www.monitor.co.ug/News/Education/-girls-science-related- highest in world. Retrieved from https://observer.ug/education/85-
courses--Physics-chemical-equations/688336-4719382- education/5599-uganda-teacher-absenteeism-is-highest-in-world.
7l73unz/index.html. Tashakkori, A., and Teddlie, C. (2010). Sage handbook of mixed
MoES (2016). Gender in Education Sector Policy. Retrieved from methods in social & behavioral research. Sage.
https://www.education.go.ug/files/downloads/GENDER%20IN%20ED U.S. Department of Health and Human Services (1979). The Belmont
UCATION%20SECTOR%20POLICY.pdf. Report. Accessed from http://www.hhs.gov/ohrp/regulations-and-
Muola, J. M. (2014). The effect of academic achievement Motivation policy/belmont-report/.
and home environment on academic performance among standard Wilberg, S., and Lynn, R. (1999). Sex differences in historical
eight Pupils (Doctoral dissertation). knowledge and school grades: A 26 nation study. Personality and
Musaazi, J. C. S. (2006). Educational planning: Principles, tools and Individual Differences, 27(6): 1221-1229.
applications in the developing World. Makerere University Printery. Zhang, M. (2003). Links between school absenteeism and child
Nielsen, H. S. (1998). Child labor and school attendance: Two joint poverty. Pastoral Care in Education, 21(1): 10-17.
decisions. University of Aarhus CLS Working Paper, (98-015).
Nosek, B. A., and Smyth, F. L. (2011). Implicit social cognitions predict
sex differences in math engagement and achievement. American
Educational Research Journal, 48(5): 1125-1156
Nsubuga, Y. (2008). Impact of leadership styles on student
performance in sciences in secondary schools in Uganda.
Unpublished PhD dissertation. Nelson Mandela Metropolitan
University. Port-Elizabeth, South Africa.
Onwuegbuzie, A. J., and Leech, N. L. (2007). A call for qualitative
power analyses. Quality and Quantity, 41(1): 105-121.
Patel, Y., and Dexter, S. (2014, March). Using multiple representations
to build conceptual understanding in science and mathematics. In
Society for Information Technology & Teacher Education
International Conference (pp. 1304-1309). Association for the
Advancement of Computing in Education (AACE).
Petersen, J., and Hyde, J. S. (2014). Gender-related academic and
occupational interests and goals. Advances in Child Development
and Behavior, 47: 43-76.
Privault, C., O'neill, J., Renders, J. M., Ciriza, V., and Hoppenot, Y.
(2012). U.S. Patent No. 8,165,974. Washington, DC: U.S. Patent and
Trademark Office.
Romm, N. R. (2018). Responsible research practice: Revisiting
transformative paradigm in social research. Springer.
Sahragard, R., Baharloo, A., and Soozandehfar, S. M. A. (2011). A
closer look at the relationship between academic achievement and
language proficiency among Iranian EFL students. Theory and
Practice in Language Studies, 1(12): 1740-1748.
Schaefer, G. O., and Wertheimer, A. (2010). The right to withdraw
from research. Kennedy Institute of Ethics Journal, 20(4): 329-352.
Sekiwu, D. (2013). Integration of Values in Management of School
Discipline in Ugandan Secondary Schools. Unpublished PhD. Thesis.
Nelson Mandela Metropolitan University. Port-Elizabeth, South Africa. Citation: Musimenta, A., Adyanga, F. A., and Sekiwu, D. (2020).
Sekiwu, D., Ssempala, F., and Frances, N. (2020). Investigating the Gender and performance disparity in mathematics: A study of
relationship between school attendance and academic performance South Western Uganda. African Educational Research Journal, 8(4):
in universal primary education: The case of Uganda. African 664-673.
Educational Research Journal, 8(2): 152-160.